"agglutinative/fusional": languages vs. individual morphemes

Gavril

Senior Member
English, USA
When people make statements such as,

"Finnish is an agglutinative language"
or
"Icelandic is a fusional/inflectional language"

it seems to me that there are at least two ways of understanding them:


1)
A significant majority (not necessarily all) of the morphemes in Finnish are agglutinative, and a significant majority of those in Icelandic are fusional. (But processes such as analogy/etc. could easily change this picture over time.)

2)
All Finnish morphology is agglutinative, and all Icelandic morphology is fusional. I.e., agglutinativity and fusionality are "structural forces" that pervade all aspects of (respectively) Finnish and Icelandic morphology.


I find #2 to be a very strange view, but I've gotten the sense (maybe wrongly) that #2 is precisely how a lot of linguists understand the terms "agglutinative" and "fusional".

Is there any evidence that justifies position #2 above, or is #1 more likely the correct position?



[Edit: Changed the example of a fusional language from "English" to "Icelandic", because the fusionality of Icelandic morphology is much more pronounced than that of modern English]
 
Last edited:
  • Gavril

    Senior Member
    English, USA
    NB:
    I think that position #2 in the previous post could (at least theoretically) be valid from a prescriptive standpoint.

    For example, there might be a language whose grammar was initially codified in such a way that all irregularities were smoothed over -- resulting in a grammar that looked maximally agglutinative to anyone who opened up a handbook on this language.

    This practice of codification might then be sustained over the years, affecting the way that changes in the language were interpreted and processed by grammarians. As a result, it could appear (at least to outside learners) that this language was "maintaining" an agglutinative character over time, and "resisting" any developments towards fusionality.

    (I suspect that there are a lot of real-world languages for which this type of codifying process has occurred, but we don't need to go into specifics here.)

    However, the above only applies to the prescriptive side of morphology, not to morphology as an informal practice or semi-conscious habit.
     
    Last edited:

    ahvalj

    Senior Member
    I think we should not forget that children learning speaking don't just memorize the rules, but try to re-analyze to a certain extent the idea of the language, so your point #2 is correct in the sense that a new speaker intuitively learns what is natural and what is not in the language, he learns how to behave in it. After all, grammars of Icelandic and Finnish, or, as extreme examples, Old Irish and Turkish, are organized around wildly different principles and their speakers must have (had) almost opposite approaches to their language techniques.

    P. S. We have here Hungarian speakers living in Slovakia and Chechia: perhaps they could share their experiences with their contrasting mother tongues?
     

    Hulalessar

    Senior Member
    English - England
    An inflectional language is one which uses affixes and/or internal changes to words to mark grammatical function. Both fusional and agglutinative languages are inflectional. In a typical fusional language a single affix will indicate several different grammatical functions. In the Latin word amo the -o indicates first person, singular, present tense, indicative and active. In a typical agglutinative language each grammatical function requiring expression has its own affix.

    Neither "fusional" nor "agglutinative" are absolute terms and both are a question of degree. If Icelandic is described as fusional and Hungarian as agglutinative it is no more than telling you how on the whole they use inflectional affixes, without necessarily suggesting that Icleandic is devoid of any aspect which could be described as agglutinative or Hungarian devoid of any aspect which could be described as fusional. As Sapir said: "Speech is too variable and too elusive to be quite safely ticketed."
     

    Gavril

    Senior Member
    English, USA
    I think we should not forget that children learning speaking don't just memorize the rules, but try to re-analyze to a certain extent the idea of the language, so your point #2 is correct in the sense that a new speaker intuitively learns what is natural and what is not in the language, he learns how to behave in it.
    To some extent, maybe. But we can't take for granted that there is a generalizable "way to behave" that encompasses all of a language's morphology (and syntax, etc.).

    After all, grammars of Icelandic and Finnish, or, as extreme examples, Old Irish and Turkish, are organized around wildly different principles
    They are?

    I'd grant that the codified grammar of Finnish has more regularity (morphologically) than the codified grammar of Icelandic, and each has certain widespread grammatical distinctions that the other doesn't. But, that doesn't add up to either language being organized around any particular principle(s) -- especially if one factors out the conscious practices of each language's grammarians.

    Maybe I'm not quite getting what you mean by "organized".
     

    AndrasBP

    Senior Member
    Hungarian
    P. S. We have here Hungarian speakers living in Slovakia and Chechia: perhaps they could share their experiences with their contrasting mother tongues?
    I grew up speaking Hungarian and Russian, but as a kid I obviously didn't think about how my languages worked, it was all intuitive.

    If you look at Hungarian conjugation, how many functions do the endings, like -k and -sz have? Certainly not one.
    :thumbsup:
     

    ahvalj

    Senior Member
    Contrast the definite Gen. Pl. of the word "fjord" in agglutinating Danish: fjord-e-ne-s (stem-Plural-definiteness-Genitive — fjord - Wiktionary) and its counterpart in fusional Icelandic: firðina (one single morpheme with even the root modified comparing with the vocabulary form fjörðurinnfjörður - Wiktionary). Of course, there are no strict boundaries between language types (as almost always in the real life), but polar types of languages (or sometimes just parts of the speech: as the noun and verb may evolve differently, cp. the rather agglutinating Romance verb) represent quite dissimilar philosophies of the grammar.

    The Hungarian verbal -k has two functions: indefinite Sg. 1 (an idiosyncratic case) and the Plural (pervasive across Hungarian morphology). By the way, no Indo-European language ever invented a Plural marker common for the Plural forms of the verb, less so for the entire morphology. -Sz only marks the indefinire Sg. 2 in a part of stems.
     

    Hulalessar

    Senior Member
    English - England
    If you look at Hungarian conjugation, how many functions do the endings, like -k and -sz have? Certainly not one.
    Whilst many of the forms end in -k, the endings (i.e. inflections) are different, e.g.

    én látok
    te látsz
    ő lát
    mi látunk
    ti láttok
    ők látnak

    Saying that in a typical agglutinative language each grammatical function requiring expression has its own affix should not be taking as implying that the same suffix cannot have more than one function.
     

    muhahaa

    Member
    Finnish
    Whilst many of the forms end in -k, the endings (i.e. inflections) are different, e.g.

    én látok
    te látsz
    ő lát
    mi látunk
    ti láttok
    ők látnak

    Saying that in a typical agglutinative language each grammatical function requiring expression has its own affix should not be taking as implying that the same suffix cannot have more than one function.
    The "-sz" suffix for example, is indicative mood, present tense, indefinite, second person, singular. If you change any of these, the suffix is not -sz-. Agglutinative is at least supposed to mean that each grammatical category (mood, tense, person, definiteness) has its own invariable suffix (with the exception of phonologically controlled allomorphy).
     

    Hulalessar

    Senior Member
    English - England
    The "-sz" suffix for example, is indicative mood, present tense, indefinite, second person, singular. If you change any of these, the suffix is not -sz-. Agglutinative is at least supposed to mean that each grammatical category (mood, tense, person, definiteness) has its own invariable suffix (with the exception of phonologically controlled allomorphy).
    Even if in some languages it is the case, I do not think that it is a requirement that there is a one-to-one correspondence between affix and grammatical category. The essential feature of agglutination is that each grammatical category required to be expressed has an affix which does not merge with any other suffix.
     

    iezik

    Senior Member
    Slovenian
    2)
    All Finnish morphology is agglutinative, and all Icelandic morphology is fusional. I.e., agglutinativity and fusionality are "structural forces" that pervade all aspects of (respectively) Finnish and Icelandic morphology.

    I find #2 to be a very strange view, but I've gotten the sense (maybe wrongly) that #2 is precisely how a lot of linguists understand the terms "agglutinative" and "fusional".

    Is there any evidence that justifies position #2 above, or is #1 more likely the correct position?
    I also often find expressed opinion #2. It's not my favourite opinion. This or similar opinions are rather common in theoretical linguistics that tries to parametrize the languages with few parameters, in this case for the form of inflection. The form that you quoted seems to be held as "belief of students just studying their first linguistics subject and happily sharing knowledge in internet". Ups, maybe this is too harsh, but I expect the experienced linguists to be more precise.

    I also wouldn't like to speak about "correct" position. This reminds me of discussions in another area, physics, whether light is more correctly described as particle or a wave. I prefer to speak about "useful" concepts. The words like structural forces or pervade are for me from a science-selling jargon.

    Then, from my experience coding inflections in few languages, some sets of affixes (inflections) in Hungarian are marked to be agglutinative and some are fusional. In neighbouring languages, Slovene and Italian, no inflection is marked as agglutinative. However, the adjective superlative in Slovenian uses prefix naj- that could be regarded as agglutinative, but I treated superlative forms as using yet another root, e.g. dober-, dobr-a, boljš-i, najboljš-i. In Italian, clitics after infinitive are yet another device to produce forms, say pagar-e, pagar-ti, pagar-ti-la, what - as already mentioned - could be regarded as agglutination.

    I haven't tried all official languages of EU, but it seems to me that the written forms for all the languages can be generated by listing, (for each lexeme), one or more roots, and having fusional inflections, agglutinative inflections, semitic inflections, plus a set of clitics. And, there are probably other equivalent sets of recorded data about a language.
     
    < Previous | Next >
    Top